Material Properties and Mechanical Performances of Manufactured Factory-Produced Glass Fiber-Reinforced Autoclaved Aerated Concrete Panel

Abstract

Autoclaved aerated concrete (AAC) has gained widespread acceptance in construction as a lightweight solution for exterior and interior walls. However, traditional steel-reinforced autoclaved aerated concrete (SR-AAC) has limitations, including concerns over its ductility and difficulty in cutting during installation. The steel reinforcement also has high embodied carbon that does not align with the actions in the construction section to reach carbon neutrality shortly. This study investigated the material properties and mechanical performances of factory-produced fiber-reinforced autoclaved aerated concrete (FR-AAC) panels, aiming to examine their potential as an alternative solution. Full-scale FR-AAC panels with thicknesses of 100 mm, 150 mm, and 200 mm were manufactured and tested. Some panels were down-sampled to determine the dry density, water absorption, compressive strength, and flexural strength of the material, while the mechanical performances were evaluated through static and impact loading tests. The results showed that the average dry density and absorption of the FR-AAC material are 533 kg/m³ and 63%, respectively, with compressive strengths up to 3.79 MPa and flexural strengths reaching 0.97 MPa. All six panels tested under static uniformly distributed loading exceeded the self-weight limit by a factor of 1.5, satisfying standard requirements for load-bearing capacity. However, the brittle failure modes observed in some tests raise potential health and safety concerns. In contrast, the impact tests revealed that the panels have acceptable performances with the inclusion of fibers.

1. Introduction

Autoclaved aerated concrete (AAC) is a lightweight porous building material widely used globally for the construction industry’s non-load-bearing wall panels and building blocks [1]. Its versatility enables the manufacture of various products [2], including blocks, panels, thermal insulation materials, enclosure filler structures, and building wall insulation materials of different sizes [3,4]. AAC has primarily been employed as a sustainable alternative to traditional sintered clay bricks [5], offering advantages such as lightweight construction, good thermal insulation, exceptional sound absorption capabilities [6], machinability, and non-combustibility [7]. AAC products are manufactured from a diverse range of raw materials and can utilize industrial waste, such as fly ash [6,8], red mud [9], iron tailing [10], carbide slag [6,11], and graphite tailings [12], thereby offering low energy consumption and high environmental sustainability [13,14,15]. Furthermore, the waste AAC materials can be recycled and reused in the construction industry [16,17,18].

AAC can be categorized into two types, sand-based autoclaved aerated concrete (SAAC) and fly ash-based autoclaved aerated concrete (AAAC), depending on the raw materials used [7]. SAAC primarily utilizes low-activity siliceous materials like quartz sand, whereas AAAC is characterized by volcanic ash with high activity as its primary type [19]. The properties of AAC are also heavily influenced by the foam agents employed [20]. Aluminum powder is typically used as a gas-generating material, reacting with water and quicklime to produce hydrogen gas [21,22]. The manufacturing process typically involves several stages: mixing, pouring, foaming, static stopping of the billet, cutting, and high-temperature autoclaving [11,23,24]. Additionally, gypsum is occasionally utilized as a gas-generating regulator in manufacturing and processing AAC [22,25].

AAC exhibits a relatively low bulk density of 400–800 kg/m³, which is approximately one-third to one-sixth the density of ordinary concrete [9,26]. The complex microstructure of AAC is characterized by numerous internal pores, resulting in a void volume that can reach up to 50–90% [13,27]. This unique structure confers several benefits, including reduced self-weight and ease of handling and transportation [28]. It also exhibits low thermal conductivity, which ranges from 0.10 to 0.18 W/(m·K) [29], which is usually approximately one-quarter to one-fifth of the normal bricks [2]. A study has demonstrated that a 200 mm thick AAC wall provides thermal insulation comparable to a 490 mm thick solid clay brick wall [2]. AAC ensures that it exhibits excellent fire resistance properties. Moreover, AAC does not emit harmful gases, making it a safer material option [9].

Despite AAC’s widespread acceptance in the construction industry due to its numerous benefits as a green material, several limitations still require attention. The unique porous structure and lightweight nature can contribute to issues such as low compressive and flexural strengths, high water absorption [30], and inadequate impact resistance [1,13,31]. Consequently, research focused on enhancing the properties of AAC to better align with its application aspects in building construction will have a broader developmental trajectory.

Incorporating fibers into composite materials can significantly improve bonding between raw constituents, thereby inhibiting material cracking and crack propagation [32,33]. As such, adding fibers to the AAC matrix is a straightforward and effective means of enhancing its impact resistance and flexural strength while reducing cracking propensity. However, the preparation process of AAC requires high-temperature and high-pressure conditions, making it crucial to select suitable fibers. Not all fibers are compatible with AAC. For instance, carbon fibers exhibit poor adhesion properties and are expensive [34], polypropylene fibers may melt at elevated temperatures [35], and steel fibers can segregate in the raw material slurry, hindering uniform dispersion [33]. Additionally, the length of fibers greatly impacts the mechanical properties [35]. Longer fibers may hinder the mixing process for AAC and the clumping during the slurry pouring process may affect the uniform distribution of the final product [36]. In contrast, shorter fibers may impede pore generation during the AAC gassing process or offer limited improvement compared to the original AAC [37]. To optimize the mechanical strength of AAC materials, further investigation is necessary to determine the optimal type of fiber addition and its impact on AAC performance.

Currently, there are still many disadvantages of steel-reinforced autoclaved aerated concrete (SRAAC) panels, especially their questionable durability. Other reinforcing methods, such as non-corrosion rebars or externally bonded strengthening materials, are either too expensive or too labor-intensive. The manufacturing process of fiber-reinforced autoclaved aerated concrete (FR-AAC) will have minimal impact on the existing manufacturing solution of AAC, thus having cost-effectiveness over other alternatives. FR-AAC shall meet the performance of SRAAC and the regulatory requirements in practical applications to be considered as a viable solution, which can significantly reduce the difficulty of preparation, improve production efficiency, reduce the production cost, and be more environmentally friendly to satisfy the actual building application and ensure construction safety. Therefore, in this study, fiber-reinforced autoclaved aerated concrete panels of different sizes were prepared using mixed fibers, and the material properties (dry density, water absorption, compressive strength, and flexural strength) and mechanical performances (static loading test and impact test) were tested, aiming to investigate their potential as an alternative solution.

2. Materials and Methods

2.1. Mix Design

The material mix design for FR-AAC is presented in

Table 1. Mix design of FR-AAC.

Raw Materials	Mix Design (kg/m3)
P·II 42.5 Portland cement	160
Quicklime	200
Aluminum powder	0.6
Gypsum	40
Silica sand	396.8
Glass fiber	4

. The cement and quicklime are supposed to provide silicate and calcium, while the aluminum powder is used as an aeration agent. Silica sand was used to form an aggregate skeleton, while gypsum was used as a gas-generating regulator. The supplier provided the mix proportion based on industrial experience and practice. The produced panel has dimensions of 100 mm, 150 mm, and 200 mm in thickness and 600 mm in height. To facilitate laboratory transportation, the panels were cut into strips ranging from 2.3 m to 3 m in length. Glass fibers with a dosage of 4 kg per cubic meter (around 0.17%) are used, as shown in Figure 1. The length and diameter of fibers are around 6 mm and 200 µm, respectively. During the manufacturing process, a 4-h 120 °C steam curing at 1.0 MPa was applied, followed by a drying stage that lasted for eight hours.

2.2. Specimen Preparation

To prepare the material specimens, three samples from each of the three thicknesses were carefully selected and cut into 250 mm and 350 mm lengths, as depicted in Figure 2. This cutting arrangement considers the aeration direction and relative position to ensure that the resulting specimens represent the material’s properties.

2.3. Dry Density and Water Absorption

The dry density, moisture content, and water absorption of the material were determined according to Chinese Standard [38]. The dimensions of samples are first measured with an accuracy of 0.1 mm. The mass ($M$) of the test block is then recorded to an accuracy of 1 g. The samples were placed in an electric oven and dried at 65 °C for 24 h, followed by a temperature increase to 85 °C for another 24 h. Finally, it was adjusted to 105 °C until it reached a constant quality ($M_0$). After cooling at room temperature for 6 h, the samples were submerged in water (20 ± 2 °C) for 24 h. Following soaking, samples were gently wiped dry with a damp cloth until no visible water droplets remained on their surfaces and then weighed to an accuracy of 1 g as $M_g$. The dry density ($r_0$), moisture content ($W_s)$, and water absorption of samples ($W_r$) can be calculated via Equations (1)–(3), as follows:

$$r_0={\displaystyle \frac{M_0}{V}}\times {10}^6$$

(1)

where $r_0$ is the dry density in kg/m³, $M_0$ is the mass of the specimen after drying in g, and $V$ is the volume of the specimen in mm³.

$$W_s={\displaystyle \frac{{M-M}_0}{M_0}}\times 100\%$$

(2)

where $W_s$ is the mass moisture content in % and $M$ is the mass of the specimen before drying in g.

$$W_r={\displaystyle \frac{{M_g-M}_0}{M_0}}\times 100\%$$

(3)

where $W_r$ is the mass water absorption in % and $M_g$ is the mass of the specimen after water absorption in g.

2.4. Compressive Strength Tests

Before conducting the compressive strength test, the length and width of the samples were measured with electronic vernier calipers to an accuracy of 0.1 mm, allowing for the calculation of the loading area. A 300 kN universal testing machine was employed to perform tensile testing on the samples, utilizing stress control at a loading rate of 2.0 kN/s. The test was conducted with a stress rate of 0.2 MPa/s until the specimen failed, and the failure load was recorded.

2.5. Flexural Strength Tests

Three samples, with dimensions of 400 mm × 100 mm × 100 mm, were tested for flexural strength. The width and height of the blocks were measured in the middle of the test block to the nearest 0.1 mm. The specimens were placed on the bending support rollers at a distance of 300 mm between the support points, as shown in Figure 3. A 100 kN universal testing machine was used with a displacement-controlled loading rate of 1.0 mm/min until the specimens were broken. The failure load and damage location were recorded.

The flexural strength can be calculated based on the following Equation (4):

$$f_f={\displaystyle \frac{P·L}{b·h^2}}$$

(4)

where $f_f$ is the flexural strength of the block in MPa, $P$ is the destructive load in N, $L$ is the support spacing or span in mm, $b$ is the width of the specimen at 100 mm, and$h$ is the thickness of the specimen at 100 mm.

3. Material Properties

3.1. Dry Densities and Water Absorptions

The dry density, moisture content, and water absorption test results are presented in

Table 2. Dry density and water absorption tests.

Thickness (mm)	Dry Density (r0), kg/m3		Mass Moisture Content (Ws), %		Absorption (Wr), %
	Average	SD	Average	SD	Average	SD
100	526.96	3.30	23.91%	1.05%	67.19%	0.66%
150	534.18	4.77	28.45%	4.36%	60.13%	3.85%
200	537.81	2.77	25.84%	0.71%	62.62%	1.71%

. The results showed that there are no significant differences between the materials obtained from panels with different thicknesses. The average dry density and absorption of the FR-AAC material are 533 kg/m³ and 63%, respectively.

3.2. Compressive Strengths

The representative failure modes of samples cut from panels with different thicknesses can be found in Figure 4. The failure patterns of the specimens are mostly semi-explosive and are similar to that of normal concrete. There are no significant differences between specimens obtained from panels with different thicknesses. The failure patterns satisfied the standard requirements and were thus regarded as valid for all samples. The summary can be found in

Table 3. Compressive strengths.

Thickness	Compressive Strength (MPa)	Mean (MPa)	SD (MPa)
100	3.70	3.95	0.59
	4.62
	3.53
150	3.92	4.23	0.30
	4.25
	4.52
200	3.39	3.19	0.17
	3.10
	3.08

, with the results shown in Figure 5. A one-way ANOVA analysis yielded a p-value of 0.04225, meaning a significant difference among different thicknesses. The specimens obtained from panels with 100 mm and 150 mm thickness reached the compressive strengths of 3.95 MPa and 4.23 MPa, respectively, at similar levels. By contrast, the compressive strength of specimens cut from the 200 mm thick panel is relatively low at 3.19 MPa. It could be caused by the shorter high temperature curing during the autoclave process.

3.3. Flexural Strengths

The failure patterns of specimens under the flexural tests are shown in Figure 6. The load kept increasing until sudden failure occurred. The crack opened at the location close to the middle span and such patterns are similar for specimens obtained from panels with different thicknesses. The crack propagation cannot be observed before the abrupt fracture of the specimen. The fibers added to the material do not provide sufficient bridging effects in such a testing scheme. The results of the flexural strengths are shown in

Table 4. Flexural strengths.

Thickness	Flexural Strength (MPa)	Mean (MPa)	SD (MPa)
100	1.02	1.02	0.07
	0.95
	1.08
150	1.08	0.98	0.15
	1.05
	0.81
200	0.88	0.91	0.03
	0.94
	0.90

and Figure 7. A one-way ANOVA analysis led to a p-value of 0.4 due to the larger variations in data in a group with 150 mm thickness, which failed due to the existence of defects of large voids in the mid-span. The relatively smaller strength of specimens with 200 mm thickness can be attributed to the shorter high-temperature duration during the autoclave process.

The relatively smaller compressive and flexural strengths of samples obtained from the 200 mm thick panels require further investigation to ensure reliable and safe use of the proposed F-AAC material.

3.4. Scanning Electron Microscope Analyses of Material after the Flexural Strength Tests

Scanning Electron Microscope (SEM) analyses were conducted on specimens after the flexural strength tests to observe the failure patterns of the fibers and the interface between the fibers and matrices. As shown in Figure 8, a layer of hydrated products was observed on the glass fibers, as shown in Figure 8a. There are no signs of debonding after the failure, as shown in Figure 8b.

4. Static Loading

4.1. Test Setup

A minimum length of 2 m is required for the test panel. The FR-AAC panel is supported by parallel bearings whose length exceeds the width of the panel. Specifically, one support features a fixed hinge and the other a rolling hinge, as shown in Figure 9a. The panel’s displacement was measured by four dial gauges placed close to the middle of the supported edges and edges of both sides of the midspan, as shown in Figure 9b.

Six panels were tested with two specimens for each of typical 100 mm, 150 mm, and 200 mm thicknesses. Before loading, the FR-AAC panel requires weighing and recording its self-weight. A minimum of five loading levels were imposed, with each level not exceeding 30% of the panel’s self-weight. The weighted sandbags and concrete loading samples are stacked from both ends toward the middle and evenly distributed on the panel to ensure uniform loading. The displacement readings at each of the first four loading levels should be taken right after the 2-min resting period after applying the load. The fifth and final loading levels reach beyond 150% of the panel’s self-weight and a 5-min resting period shall be ensured before continuing loads until the panel fractures or breaks. The sum of loads from the initial loading level to the one preceding fracture damage was recorded as the final test result. This outcome allows for calculating the panel’s bending strength, which is the indicator for load-carrying capacity as a factor of self-weight. A load-deflection curve can also be drawn based on these findings. Per design requirements for the internal partition wall [38], the bending load of lightweight wall panels must be higher than 1.5 times the self-weight of the panel to comply with specifications.

The bending strength of FR-AAC panels under uniformly distributed loads was investigated using a supported beam model. The calculation of the maximum bending moment in the mid-span region involved several key parameters: $M=q{l}^2/8$, where $q$ was the uniform load applied per unit length (N/mm) and $l$ represented the span of the plate (mm). The maximum normal stress can be equated with the maximum moment $M_{max}$. Consequently, the ultimate equation for calculating the bending strength of a rectangular section of AAC panels reinforced with fibers is shown in Equation (5)

$${\sigma }_{max}={\displaystyle \frac{M_{max}}{\frac{b_p{h_p}^2}{6}}}={\displaystyle \frac{3}{4}}·{\displaystyle \frac{q}{b_p}}·{\displaystyle \frac{l^2}{{h_p}^2}},$$

(5)

where $b_p$ is the width of the panel in mm and $h_p$ is the thickness of the panel in mm.

4.2. Testing Results

The load versus net deflection curves (middle span deflection minus the deflection at the support) are shown in Figure 10a. The panels behave mostly linearly, as shown in the plots, until the sudden failure, which usually leads to catastrophic breakage of the panel into two or more pieces, as shown in Figure 10b. The two specimens cut from panels with the same thickness performed similarly, except the first specimen cut from the 150 mm thick panel “150 mm⁻¹”, which failed much earlier than its counterpart “150 mm⁻²”. Further exploration led to the observation of a large defect, as shown in Figure 10c, which is regarded as the primary reason for such premature failure. Larger voids can be observed in the center of the defect area, while fewer small voids were distributed in the rest. The homogeneity of material was probably affected by either the bad mix of raw material or the agglomerated raw material, such as cement or quicklime. Such defects were only observed on the fracture surface of specimen “150 mm⁻¹”. The average flexural strength (excluding specimen “150 mm⁻¹”) is 0.60 MPa, and more details can be found in

Table 5. The static loading test results.

Specimen	100 mm−1	100 mm−2	150 mm−1	150 mm−2	200 mm−1	200 mm−2
Self-weight (kg)	109	115	156	157	262	264
Span (mm)	2270	2270	2370	2370	2900	2900
Peak load (kN)	2.55	2.06	2.89	4.66	5.29	6.32
Load to self-weight ratio (-)	2.39	1.83	1.89	3.03	2.06	2.44
Ultimate moment (kN·m)	0.72	0.58	0.86	1.38	1.92	2.29
Flexural strength (MPa)	0.72	0.58	0.38	0.61	0.48	0.57

. It shall be noted that more specimens are required to obtain statistically meaningful results. Hence, the testing results of individual specimens are reported in Table 5.

Due to the limited available number of specimens (two for each thickness) for the flexural testing, more checking was enforced to evaluate the consistency of the material. Concerning the defects, three sectional cuts were enforced on the three different panel thicknesses, which ended up with six cross-sections, as shown in Figure 11.

A brief visual inspection did not find defects in equivalent size, as shown in Figure 10c. Some detects with small sizes can be identified from panels, especially the ones with a thickness of 150 mm. It is assumed that these defects are mainly due to the impurities of the raw material, such as silica sand and hydrated lime. Further sieving of the raw material might be helpful to reduce the probability of the existence of defects.

4.3. Comparison with Steel-Reinforced AAC Panels (SR-AAC)

The load capacities and equivalent flexural strengths of steel-reinforced AAC panels were estimated for the panel thickness of 100, 150 and 200 mm. It shall be noted that for internal wall slabs with a thickness less than or equal to 150 mm in thickness and with a length less than 3000 mm, single-layer reinforcement mesh is usually adopted. Otherwise, double reinforcement shall be enforced. Moreover, there shall be no less than 4 reinforcements in the longitudinal direction, and their diameter shall be larger than or equal to 4 mm. Based on the regulation, typical reinforcement schemes are shown in

Table 6. Reinforcement schemes for SR-AAC panel.

Panel Thickness	Reinforcement Diagram
100 mm
150 mm
200 mm

.

Single-reinforced panels are not expected to exhibit significant enhancements in flexural resistance. The primary benefit of adding a middle reinforcement layer is to increase ductility and prevent brittle failure. For double-reinforced SR-AAC, assuming full utilization of the reinforcement, the following Equation (6) can be used to estimate the flexural strength:

$${\sigma }_{eq}={\displaystyle \frac{6·f_{sk}·A_s·n·h_0}{b_p·{h_p}^2}},$$

(6)

where $f_{sk}$ is the standard value of the tensile strength of the reinforcement in MPa, $A_s$ is the surface area of the steel rebar in mm², $n$ is the number of longitudinal steel rebar, and $h_0$ is the centre-to-centre distance between the top and bottom layers of steel rebars in mm. By using HPB300 steel reinforcement ($f_y=270 \mathrm{M}\mathrm{P}\mathrm{a}$) purchased from the local supplier in Suzhou, China. Two layers of four rebars, as shown in Table 6. with the 150 mm rebar layer distance, the theoretical equivalent flexural strength is 0.51 MPa, which is close to the level calculated based on the test results for FR-AAC with 200 mm thickness (0.48 MPa and 0.57 MPa, respectively, for the two tested specimens). It shall be noted that the estimation according to Equation (6) ignored the contribution of AAC itself, which underestimated the capacity of SR-AAC. Experiment verification is still necessary. Moreover, a comprehensive comparison between the two types of materials shall also cover other aspects, such as ductility and durability, in order to justify the potential replacements.

5. Impact Tests

5.1. Single Panel Tests

The exterior and interior wall panels are also subjected to impact loading, such as those caused by projectiles due to extreme wind speeds. To check the impact resistance of the panel made from FR-AAC, the fixture, as shown in Figure 12a,b, was configured. One panel with a thickness of 100 mm was tested. A stainless steel ball weighing 1.8 kg is designed to fall freely from heights of 0.5 m, 1.0 m, and 1.5 m, respectively, striking the panel at the impact point, as shown in Figure 12a. The impacts were repeated three times at the same time before moving to the next height level if the panel was intact after each impact loading.

The DH5922 dynamic signal test and analysis system was utilized to record the accelerations during the impact. As shown in Figure 12a, three piezoelectric accelerometers were installed, and the sampling frequency was set to 1 kHz. The accelerometer has a range of 10 g, a resolution of 0.00029 g, and an axial sensitivity of 100.2 mV/g, which is suitable for impact tests. After each impact, the panel was inspected for detecting cracks before the next impact.

5.2. Assembled Wall Tests

The impact testing of the assembled wall shall be configured based on standard requirements [39], as shown in Figure 13a, which consists of three panels placed side by side and glued together. The structural adhesive was applied to the sides of the panels and then the panels were pushed side by side together to form the testing wall. A 24-h curing time was allowed before applying the impacting. The assembled wall panel was supported by the bottom steel rod and was clamped at the top and bottom to prevent lateral movement. The impact was induced by the sandbags that weigh 30 kg and freely swing along a fixed rope from a specified height of 1.5 m. Therefore, the point of impact is close to half the height of the wall, with an impulse impact lasting for less than one-tenth of a second. The same accelerometers were attached to the top and back of the assembled wall, as shown in Figure 11b,c, to monitor and record the responses of the wall during impact.

5.3. Results from the Single Panel Tests

The panel sustained nine impacts before fracturing into pieces, as shown in Figure 14a. A typical failure mode of the panel can be found in Figure 14a, while the representative responses of the accelerometer can be found in Figure 14b, c, and d, respectively, for sensors A1, A2, and A3, under the first impact of sandbag dropping from a 0.5 m height.

To check the potential damage of the panel, shock response spectra (SRS) were generated based on the records of acceleration, as shown in Figure 15a–c, for a 0.5 m drop, 1.0 m drop, and 1.5 m drop, respectively. It was found that A1 and A2 had similar responses, while the accelerations recorded by A3 were smaller than those of A1 and A2 because it is away from the impact location. The SRS values from the first few impacts are very close to each other, for all three sensors, and reflect minimal damage after those impact actions. However, a significant response reduction was found in the 1.5 m drop height group, for which the response of the second impact is much smaller, revealing accumulated internal damage to the panel.

5.4. Assembled Wall Impact Test Results

The assembled wall sustained two successive impacts while fractured during the third. The failure mode is shown in Figure 16a. A longitudinal crack at approximately the middle height propagated across all three panels, which is more visible from the back view, as shown in Figure 16b. The wall thus lost its stability and fractured into two pieces. The side view of the wall, as shown in Figure 16c, clearly shows how the crack propagated from the backside to the front side.

The representative acceleration and strain recordings stored during the first impact are shown in Figure 17. The acceleration records reveal that following the impact, the specimen’s acceleration rapidly reaches a peak value of 1.8 m/s² between 2 ms and 5 ms. Notably, this peak is followed by a rapid decay, and a second peak occurs after the data reach their maximum due to the secondary impact due to the lack of restraint system of the sandbag swing. However, the impact energy of such a secondary impact is very small; thus, its impact can be neglected.

Shock response spectra were also generated, as shown in Figure 18. It was observed that the response from the first impact is much less than the second and third impacts, which may be attributed to the fact that the sandbags will be more consolidated after the first impact. In other words, the load is more evenly distributed in the first impact through the length of the sandbag. A slight decrease in the response in the third impact from the second impact was observed, regarded as a sign of degradation that indicates cumulative damage due to prior impact actions.

Although the assembled wall sustained the first two impacts, which showed the effectiveness of the added fibers and thus satisfied the minimal requirement for the partition wall in the buildings [39], considerable damage was induced. Currently, no standards exist for cumulative damaging testing, and further study is necessary.

6. Conclusions and Remarks

This study presents the experiment investigations of FACC panels, including the material property testing and static and impact testing of panels. Conclusions can be drawn as follows.

• The average dry density and absorption of the FR-AAC material are 533 kg/m³ and 63%, respectively, while the compressive strength and flexural strength vary with the panel thickness. The compressive and flexural strengths are smaller for the samples taken from panels with 200 mm thickness;
• All six panels tested under static uniformly distributed loading exceeded 1.5 times the self-weight limit, which satisfies the standard requirement. However, the failure is brittle and the panel is sensitive to defects that exist in the critical region. Further investigations on ductility and durability are necessary before its implementation;
• The wall panels passed the impact test and thus satisfied the requirement for its application as building partition walls. However, it is evident that considerable levels of damage were induced through the impact, and the wall failed during the third repeating impact. Its application in more severe conditions, such as those for exterior walls, shall be further verified;

Besides the conclusions, several remarks and future directions of investigations are listed as follows;

• The durability of FRAAC is expected to be better than SRAAC, mainly due to the better durability of the fibers than the steel rebars. However, more long-term tests shall be required to compare the two solutions, especially under the alternation of moisture and temperature, which could be more critical to the interface between fibers and the AAC matrix. Such degradation is worth further exploration;
• The FACC and the wall panels demonstrated satisfactory performances regarding their mechanical strengths. However, more verifying work shall be conducted to ensure such products’ reliable and safe applications. Besides durability, health and safety during construction shall also be addressed, considering the brittle nature of the F-AAC panels. In order to overcome such shortcomings, it is suggested to use mixed fibers of different lengths or a mixture of several types of fibers. The type and dosage of such fiber mixture shall be further investigated;
• Due to the limitation of the availability of materials and panels manufactured in the factory environment, the fiber type and content were pre-selected based on the manufacturer’s recommendation. Future optimization will be conducted in the laboratory at a smaller scale. The final selection shall be based on the overall consideration of the mechanical properties and the sustainability of materials, such as their embodied carbon and energy levels. Fibers with less environmental burden, such as natural fibers or fibers produced from waste materials, can be considered to enhance their sustainability further;
• The sensitivity of material with respect to the thickness of the panel, more specifically, the impacts from the different durations at the designated curing temperature, is worth further investigation. The defects within the material are critical, and the appropriate quality control process is worth further investigation;
• The sustainability benefits of using FR-AAC over SRAAC shall be further quantified using the life cycle assessment method. The embedded carbon of FR-AAC can be further reduced by utilizing waste solid material [40,41] or introducing carbon dioxide curing [42], which has been approved by prior research.